Linguistically-driven automated text formatting

Abstract

Systems and techniques for linguistically-driven automated text formatting are described herein. Data representing the linguistic structure of input text may be received from Natural Language Processing (NLP) Services, including but not limited to constituents, dependencies, and coreference relationships. A text model of the input text may be built using the linguistic components and relationships. Cascade rules may be applied to the text model to generate a cascaded text data structure. Cascaded data may be displayed on a range of media, including a phone, tablet, laptop, monitor, VR/AR devices. Cascaded data may be presented in dual screen formats to promote more accurate and efficient reading comprehension, greater ease in teaching native and foreign language grammatical structures, and tools for remediation of reading-related disabilities.

What is claimed is: 
     
       1. A non-transitory computer-readable medium comprising instructions for cascaded text formatting, when executed by at least one processor of a computing device, cause the at least one processor to perform operations that:
 receive a text portion to be cascaded; 
 invoke a trained model to process the text portion, the trained model to receive the text portion as input and to identify formatting changes to create cascaded text from the text portion, 
 wherein the formatting changes includes respective line breaks and respective horizontal displacements to be added among respective words of the text portion, 
 wherein the respective line breaks are added to the text portion based on grammatical units of the respective words, wherein the grammatical units comprise constituency, and wherein the respective line breaks establish a vertical position of the respective words among separate lines or continuous units, and 
 wherein the respective horizontal displacements are added to the text portion based on syntactic functions of the respective words, wherein the syntactic functions comprise grammatical dependencies, and wherein the respective horizontal displacements establish a horizontal position of the respective words among the separate lines or continuous units; and 
 cause output of the cascaded text in real-time based on the formatting changes. 
 
     
     
       2. The non-transitory computer-readable medium of  claim 1 , wherein the trained model is executed at the computing device, and wherein the trained model outputs the cascaded text or a language model to generate the cascaded text. 
     
     
       3. The non-transitory computer-readable medium of  claim 1 , wherein the trained model is operated at a networked-accessible service at a computer system located remote to the computing device, and wherein the networked-accessible service outputs the cascaded text or a language model to generate the cascaded text. 
     
     
       4. The non-transitory computer-readable medium of  claim 1 , further comprising instructions that cause the at least one processor to perform operations that:
 output the cascaded text on a user interface of the computing device, wherein the user interface provides at least one display modification to the cascaded text including: a split-screen, automatic scrolling, text coloring, text italics, text underlining, or text highlighting. 
 
     
     
       5. The non-transitory computer-readable medium of  claim 1 , wherein the grammatical units of the respective words are based on constituency rules, and wherein the syntactic functions of the respective words are based on dependency rules. 
     
     
       6. The non-transitory computer-readable medium of  claim 1 , wherein the trained model implements at least one of: a neural network, a machine learning classifier, a natural language processing (NLP) parser, a rule engine, or an inference engine. 
     
     
       7. The non-transitory computer-readable medium of  claim 1 , wherein the trained model is trained from a plurality of cascaded text examples corresponding to the plurality of cascaded text examples. 
     
     
       8. A method for cascaded text formatting, performed by a computing device, comprising:
 identifying a text portion to be cascaded; 
 processing the text portion with a trained model, the trained model configured to receive the text portion as input and to identify formatting changes to create cascaded text from the text portion, 
 wherein the formatting changes includes respective line breaks and respective horizontal displacements to be added among respective words of the text portion, wherein the respective line breaks are added to the text portion based on grammatical units of the respective words, wherein the grammatical units comprise constituency, and wherein the respective line breaks establish a vertical position of the respective words among separate lines or continuous units, and 
 wherein the respective horizontal displacements are added to the text portion based on syntactic functions of the respective words, wherein the syntactic functions comprise grammatical dependencies, and wherein the respective horizontal displacements establish a horizontal position of the respective words among the separate lines or continuous units; and 
 displaying the cascaded text in real-time based on the formatting changes. 
 
     
     
       9. The method of  claim 8 , wherein the trained model is executed at the computing device, and wherein the trained model outputs the cascaded text or a language model to generate the cascaded text. 
     
     
       10. The method of  claim 8 , wherein the trained model is operated at a networked-accessible service at a computer system located remote to the computing device, and wherein the networked-accessible service outputs the cascaded text or a language model to generate the cascaded text. 
     
     
       11. The method of  claim 8 , further comprising:
 outputting the cascaded text on a user interface of the computing device, wherein the user interface provides at least one display modification to the cascaded text including: a split-screen, automatic scrolling, text coloring, text italics, text underlining, or text highlighting. 
 
     
     
       12. The method of  claim 8 , wherein the grammatical roles units of the respective words are based on constituency rules, and wherein the syntactic functions of the respective words are based on dependency rules. 
     
     
       13. The method of  claim 8 , wherein the trained model implements at least one of: a neural network, a machine learning classifier, a natural language processing (NLP) parser, a rule engine, or an inference engine. 
     
     
       14. The method of  claim 8 , wherein the trained model is trained from a plurality of cascaded text examples corresponding to the plurality of cascaded text examples. 
     
     
       15. An apparatus, comprising:
 means for obtaining a text portion to be cascaded; 
 means for invoking an automated formatting system to process the text portion, the automated formatting system to receive the text portion as input and to identify formatting changes to create cascaded text from the text portion, 
 wherein the formatting changes includes respective line breaks and respective horizontal displacements to be added among respective words of the text portion, wherein the respective line breaks are added to the text portion based on grammatical units of the respective words, wherein the grammatical units comprise constituency, and wherein the respective line breaks establish a vertical position of the respective words among separate lines or continuous units, and 
 wherein the respective horizontal displacements are added to the text portion based on syntactic functions of the respective words, wherein the syntactic functions comprise grammatical dependencies, and wherein the respective horizontal displacements establish a horizontal position of the respective words among the separate lines or continuous units; and 
 means for outputting the cascaded text based on the formatting changes. 
 
     
     
       16. The apparatus of  claim 15 , wherein the automated formatting system outputs the cascaded text or a language model to generate the cascaded text. 
     
     
       17. The apparatus of  claim 15 , wherein the means for outputting the cascaded text provides at least one display modification to the cascaded text including: a split-screen, automatic scrolling, text coloring, text italics, text underlining, or text highlighting. 
     
     
       18. The apparatus of  claim 15 , wherein the grammatical units of the respective words are based on constituency rules, and wherein the syntactic functions of the respective words are based on dependency rules. 
     
     
       19. The apparatus of  claim 15 , wherein the automated formatting system implements at least one of: a neural network, a machine learning classifier, a natural language processing (NLP) parser, a rule engine, or an inference engine. 
     
     
       20. The apparatus of  claim 15 , wherein the automated formatting system includes a trained model that is trained from a plurality of cascaded text examples corresponding to the plurality of cascaded text examples. 
     
     
       21. A method for cascaded text formatting of an input text, comprising:
 receiving, by at least one processor of a computing device, the input text to be cascaded; 
 obtaining, by the at least one processor, dependency information for the input text; 
 determining, by the at least one processor, constituency information for the input text; 
 generating, by the at least one processor, a language model from the constituency information and the dependency information, wherein the language model is a data structure comprising: a parse tree representing dependencies among constituents of the input text, and encoded data representing relationships between the constituents and dependencies of the constituents of the input text; 
 applying, by the at least one processor, cascade rules to the constituents of the input text, using the language model, the cascade rules comprising instructions to:
 determine a vertical arrangement of each constituent, wherein each constituent comprises a word or group of words that functions as an individual grammatical unit in a hierarchy of the parse tree of the generated language model, and wherein the vertical arrangement separates the constituents of the input text into respective lines or continuous units; and 
 determine a horizontal arrangement of each constituent, wherein one or more dependency relationships and a head of each constituent are determined from the encoded data of the language model, and wherein the horizontal arrangement provides a same horizontal displacement for the constituents of the input text that have a dependency relationship to a same head or a dependency relationship on each other; 
 
 generating output, by the at least one processor, wherein the output includes line breaks to implement the vertical arrangement and indentation to implement the horizontal arrangement within the input text; and 
 causing, by the at least one processor, a display of the output on a display device. 
 
     
     
       22. A non-transitory computer-readable medium including instructions for cascaded text formatting of an input text, which when executed by at least one processor of a computing device, cause the at least one processor to perform operations that:
 receive the input text to be cascaded; 
 obtain dependency information for the input text; 
 determine constituency information for the input text; 
 generate a language model from the constituency information and the dependency information, wherein the language model is a data structure comprising: a parse tree representing dependencies among constituents of the input text, and encoded data representing relationships between the constituents and dependencies of the constituents of the input text; 
 apply cascade rules to the constituents of the input text, using the language model, the cascade rules comprising instructions to:
 determine a vertical arrangement of each constituent, wherein each constituent comprises a word or group of words that functions as an individual grammatical unit in a hierarchy of the parse tree of the generated language model, and wherein the vertical arrangement separates the constituents of the input text into respective lines or continuous units; and 
 determine a horizontal arrangement of each constituent, wherein one or more dependency relationships and a head of each constituent are determined from the encoded data of the language model, and wherein the horizontal arrangement provides a same horizontal displacement for the constituents of the input text that have a dependency relationship to a same head or a dependency relationship on each other; 
 
 generate output, wherein the output includes line breaks to implement the vertical arrangement and indentation to implement the horizontal arrangement within the input text; and 
 cause a display of the output on a display device. 
 
     
     
       23. A system for cascaded text formatting of an input text, comprising:
 at least one processor; and 
 memory including instructions that, when executed by the at least one processor, cause the at least one processor to perform operations to:
 receive the input text to be cascaded; 
 obtain dependency information for the input text; 
 determine constituency information for the input text; 
 generate a language model from the constituency information and the dependency information, wherein the language model is a data structure comprising: a parse tree representing dependencies among constituents of the input text, and encoded data representing relationships between the constituents and dependencies of the constituents of the input text; 
 apply cascade rules to the constituents of the input text, using the language model, the cascade rules comprising instructions to:
 determine a vertical arrangement of each constituent, wherein each constituent comprises a word or group of words that functions as an individual grammatical unit in a hierarchy of the parse tree of the generated language model, and wherein the vertical arrangement separates the constituents of the input text into respective lines or continuous units; and 
 determine a horizontal arrangement of each constituent, wherein one or more dependency relationships and a head of each constituent are determined from the encoded data of the language model, and wherein the horizontal arrangement provides a same horizontal displacement for the constituents of the input text that have a dependency relationship to a same head or a dependency relationship on each other; 
 
 generate output, wherein the output includes line breaks to implement the vertical arrangement and indentation to implement the horizontal arrangement within the input text; and 
 cause a display of the output on a display device. 
 
 
     
     
       24. The method of  claim 21 , wherein the horizontal arrangement provides an alignment of the respective lines that are linguistically related to each other, and wherein a dependency relationship determines the indentation for the respective lines that are linguistically dependent on each other. 
     
     
       25. The non-transitory computer-readable medium of  claim 22 , wherein the horizontal arrangement provides an alignment of the respective lines that are linguistically related to each other, and wherein a dependency relationship determines the indentation for the respective lines that are linguistically dependent on each other. 
     
     
       26. The system of  claim 23 , wherein the horizontal arrangement provides an alignment of the respective lines that are linguistically related to each other, and wherein a dependency relationship determines the indentation for the respective lines that are linguistically dependent on each other.